Explaining disparities in colorectal cancer screening among five Asian ethnic groups: A population-based study in California

Colorectal cancer (CRC) is the second leading cause of cancer deaths in the United States and among the most common cancers in all racial/ethnic subgroups, including Asian Americans. Overall, the incidence and mortality of CRC is lower in Asian Americans than in Non-Hispanic whites [1]. Only few reports on the burden of CRC disaggregate Asian American populations: California Cancer Registry data indicate that the incidence of CRC is higher in Japanese American males and females than in Non-Hispanic whites and other racial/ethnic groups and Japanese American males also have a significantly higher CRC mortality [2]. Using data from the Surveillance, Epidemiology, and End Results program, Lin and colleagues demonstrated that Filipino American males are significantly less likely than other racial/ethnic groups to be diagnosed at an early stage. In addition, their 5 year survival rate is significantly lower than those of other racial/ethnic groups. Chinese American females have poorer 5-year survival after CRC than Japanese or Filipina [3]. Population-based survey data indicate that the proportion of the population over age 50 screened for CRC significantly increased in the United States and in California between 2001 and 2005 [4, 5]. However, these gains did not accrue equally across all racial and ethnic groups. Using California Health Interview Survey (CHIS) data, we recently reported the prevalence of CRC screening (ever screened) in 2001, 2003 and 2005 in California among whites, blacks, Latinos, Asians and the population overall, and among five Asian American groups with adequate sample sizes for analysis, Chinese, Filipino, Japanese, Korean and Vietnamese Americans [5]. The prevalence of screening among Asians in aggregate was consistently 11 percentage points lower than among the overall population in all three years (58%, 59% and 62% in 2001, 2003 and 2005 versus 69%, 70% and 73%), and 16 percentage points lower than among whites (74%, 75% and 78%). However, analysis of the Asian subgroups revealed more nuanced differences. Among the subgroups, Japanese Americans had the highest prevalence, with age- and gender-standardized levels of 71%, 72% and 77% in 2001, 2003 and 2005, respectively, while Korean Americans had the lowest, with levels of 49%, 43% and 33%; the other groups were at intermediate levels. The decline among Korean Americans contrasted with rising trends for Chinese, Filipino, Japanese and Vietnamese Americans. Overall, the gap between the highest and lowest prevalence groups, Japanese and Korean Americans, increased during this relatively short time period. These findings highlight the importance of disaggregating Asian American subgroups when monitoring health indicators to avoid masking differences among them [6, 7].

Prior studies investigating differences in cancer screening utilization among Asian American subgroups have attributed these differences to underlying variations in demographic characteristics such as education and income, English proficiency and access to health care [6–10]. The purpose of this investigation was to examine the role of these factors in explaining the observed differences in CRC screening prevalence among the five Asian American ethnic subgroups, and in particular the low levels and declining prevalence of CRC screening among Korean Americans. To this end, we conducted hierarchical regression analyses progressively controlling for demographic characteristics, English proficiency and access to care, as well as subgroup comparisons on reasons for non-adherence to screening guidelines and rates of doctor recommendation of CRC screening to identify differences in these factors which might explain screening prevalence disparities.

The California Health Interview Survey (CHIS) is a population-based random digit dial telephone survey conducted biennially since 2001 [11]. The CHIS is designed to provide cross-sectional, population-based statewide estimates of health indicators for all major racial and ethnic groups as well as several Asian ethnic subgroups. The CHIS obtains oral informed consent from each respondent prior to conducting the survey. This analysis was exempt from Institutional Review Board (IRB) review by the University of California, Los Angeles IRB because it was based on existing data and information was recorded in such a manner that subjects could not be identified. Data analysis was conducted in 2009. We merged data from the 2001, 2003, and 2005 surveys and created an analysis dataset consisting of adults 50 years of age and older with no history of CRC who self-identified as Chinese (N = 1,432), Filipino (N = 753), Vietnamese (N = 709), Korean (N = 675) or Japanese American (N = 619). Since Asian Americans make up 12.5% of the population in California, which is a much larger proportion than the national average of 4.5% (http://quickfacts.census.gov/qfd/states/06000.html, accessed 12/15/2009), and since CHIS oversampled several Asian American groups, this data set is well suited to examining disparities in CRC screening among Asian American ethnic groups.

After controlling for differences in education, marital status, employment status and federal poverty level (Model 2), Korean Americans were the only group that had significantly lower screening prevalence than Japanese Americans in 2001, and whose trend to 2005 remained significantly depressed. In addition, being married and having a higher income level (≥ 300% poverty level) were associated with increased odds while being employed was associated with decreased odds of ever receiving CRC screening. Model 3 added English proficiency and Model 4 added health care access as covariates. After adding these covariates, all of which were significantly associated with receipt of CRC screening in the expected direction (higher odds of screening with higher English proficiency and health care access), screening prevalences in 2001 were no longer significantly different among the Asian subgroups. However, the trend in screening prevalence among Korean Americans from 2001 to 2005 remained significantly depressed compared to Japanese Americans. The odds ratio for difference in trend for Korean Americans was fairly stable across all models, ranging from 0.26 to 0.33, suggesting that the added covariates did little to explain the declining trend.

Our analysis shows that differences in screening prevalences between Japanese Americans and Chinese, Filipino, Korean and Vietnamese Americans in California in 2001 were no longer significant after controlling for demographic characteristics and English proficiency. However, the depressed trend in screening from 2001 to 2005 among Korean Americans compared to Japanese Americans was not explained by differences in these variables nor by differences in access to care. Korean Americans are among the more recent immigrant groups and their numbers have increased rapidly since 1965 [14]. They differ from the other Asian ethnic groups studied here in that many are small business owners with no health insurance and have a lower median household income than any other group [15]. Yet our analyses showed that differences in these correlates did not fully explain the decline. Furthermore, Korean Americans were the only subgroup that had a significantly depressed trend.

In addition to demographic and socio-economic characteristics and access to health care, health beliefs influence health care utilization [12]. Thus, the stated reasons for non-adherence with CRC screening guidelines may shed some light on the reasons for this decline. First, about one-third of Korean American respondents stated that they did not get screened because they had no health problem. This suggests that Korean (and Vietnamese) Americans are less familiar with the concept of routine screening to detect health problems before the onset of symptoms than other groups that were included in this analysis. Future programs to promote CRC screening in these Asian subgroups should stress the need for screening before symptoms develop. Second, while the proportion of respondents who reported receiving a doctor's recommendation for CRC screening increased from 2001 to 2005 among Filipino Americans and remained stable among Chinese, Japanese and Vietnamese Americans, the proportion trended towards a decrease among Korean Americans. Many studies have shown that a doctor's recommendation is one of the strongest predictors of screening among Asian Americans and other racial/ethnic groups [16–18]. Low levels of health insurance among Korean Americans [19] may explain why they are less likely to receive a doctor's recommendation for screening than other ethnic groups. In addition, Korean Americans who are not comfortable communicating in English may choose a Korean-speaking physician who may be reluctant to recommend CRC screening because of their own lack of knowledge of screening guidelines, their patients' unfamiliarity with the concept of screening and the insufficient health care system referral networks and reimbursement for screening [19]. Finally, a few studies have shown that Korean Americans have other barriers to CRC screening that have not been assessed in CHIS, including lack of transportation and cultural barriers, such as fear of being a burden to one's family if diagnosed with cancer [16]. However, the influence of general and culturally specific barriers on CRC screening among Korean Americans and other Asian populations is poorly understood.

Interestingly, when asked the main reason for non-adherence to CRC screening, very few respondents in these Asian subgroups cited cost, lack of insurance or lack of a doctor. Even after consolidating these reasons into a single category reflective of access to care, the cell counts were too low to allow us to evaluate this as a separate category in the analysis. Hence we collapsed these reasons into the "Other" category in Table 3. Instead, the most frequently cited reason for non-adherence to CRC screening was "being unaware of the test".

Correlates of CRC screening that we found in this large California population-based sample of Asian Americans are similar to those found among African Americans, Latinos, non-Hispanic whites and in other Asian American samples [8, 10, 18, 20, 21]. Odds of CRC screening increased with being married, having a higher level of income, having a higher level of acculturation, having health insurance, a usual source of care and a doctor's visit within the past year. In our sample, highest odds ratios were associated with recent doctor's visit, having a usual source of care and high acculturation. Only few studies have examined the association between employment and screening and employment has not emerged as an important correlate in the screening literature [10, 18, 20, 21]. Our finding that unemployed Korean Americans are more likely to be screened for CRC is unexpected. However, it is consistent with two prior studies among Korean Americans [8, 16]. In one of our early studies, we found that full- or part-time employment was associated with lower odds of CRC screening among Korean American women, but not among Filipino American women [8]. In a recent study we found that employed Korean Americans residing in Los Angeles were less likely to be screened than those who were unemployed [16]. In that study, 15% of respondents stated "not being able to take time off work" as a reason for not getting CRC screening, which may be one of the reasons for this unexpected correlation. Future studies should further explore this potential barrier to screening.

Differences in demographic characteristics, English proficiency and access to care explain disparities in 2001 CRC screening prevalence among Japanese, Chinese, Filipino, Korean and Vietnamese Americans residing in California, but even after controlling for these factors, a significant decline in CRC screening between 2001 and 2005 among Korean Americans as compared to Japanese Americans persists and remains unexplained. Future studies that provide a better understanding of the reasons for this decline will be crucial for developing culturally appropriate interventions to promote CRC screening. Our data also suggest a need to more fully understand the role and behavior of providers that serve the Korean community in recommending CRC screening to their Korean American patients.